Tower height bounds for nₚ(α,β) at small primes
Conjecture (Theorem 2 extends to primes p < 19) · arXiv:1708.08486
Status open high confidence
The conjecture asks whether the tower-height bounds on $n_p(\alpha,\beta)$ from Theorem 2 of arXiv:1708.08486 extend from primes $p \geq 19$ to all primes $p < 19$. The obstacle identified by the authors is that, in the large-$\varepsilon$ regime ($\varepsilon = \alpha^3 - \beta$ large), the best known upper bounds on the largest subset of $\mathbb{F}_p^n$ without a three-term arithmetic progression are not strong enough to drive the required lower bound for small primes. No subsequent paper resolving this conjecture was found across the indexed literature as of May 2026.
Reviewer notes. No follow-up resolving the conjecture for primes p < 19 was found. The 2026 paper arXiv:2605.13628 (Conlon, Fox, Pham) addresses arithmetic progressions with restricted differences via Tao's slice rank method but does not address the tower-height conjecture for small primes. Progress on cap set bounds (e.g., arXiv:2401.12802, Ellenberg-Gijswijt-type improvements) has continued, but none of the surveyed results appear to close the gap for the large-epsilon regime in the small-prime case.
Context
Theorem 2 is proved only for primes $p \geq 19$. The authors note that the upper bounds in Theorem 2 hold for all primes, and the lower bound holds when $\varepsilon = \alpha^3 - \beta$ is small. The obstacle for $p < 19$ in the large-$\varepsilon$ regime is that the best known bounds on the largest subset of $\mathbb{F}_p^n$ without a three-term arithmetic progression are not strong enough to imply the required lower bound estimate in that case.
Notes. Stated in prose immediately before Theorem 2 with the words 'We conjecture these bounds should also hold for p < 19.' No formal labelled environment.
Source paper
Popular progression differences in vector spaces II
Jacob Fox, Huy Tuan Pham · 2019-11-21
https://arxiv.org/abs/1708.08486
PDF source